Open jzpcn opened 2 years ago
regression_loss = torch.where( torch.le(regression_diff, 1.0 / 9.0), 0.5 9.0 torch.pow(regression_diff, 2), regression_diff - 0.5 / 9.0 )
why you use this method to compute reg_loss rather than the normal smooth L1 loss
regression_loss = torch.where( torch.le(regression_diff, 1.0 / 9.0), 0.5 9.0 torch.pow(regression_diff, 2), regression_diff - 0.5 / 9.0 )
why you use this method to compute reg_loss rather than the normal smooth L1 loss